Search

In light of the escalating advancements in architectural intelligence and information technology, the construction of smart cities increasingly necessitates a higher degree of precision in architectural measurements. Conventional approaches to architectural measurement, characterized by their low efficiency and protracted execution time, need to be revised to meet these burgeoning demands. To address this gap, we introduce a novel architectural image processing model that synergistically integrates Restricted Boltzmann Machines (RBMs) with Convolutional Neural Networks (CNNs) to facilitate the conversion of 2D architectural images into 3D. In the implementation phase of the model, an initial preprocessing of the architectural images is performed, followed by depth map conversion via bilateral filtering. Subsequently, minor voids in the images are rectified through a neighborhood interpolation algorithm. Finally, the preprocessed 2D images are input into the integrated model of RBMs and CNNs, realizing the 2D to 3D conversion. Experimental outcomes substantiate that this novel model attains a precision rate of 97%, and significantly outperforms comparative algorithms in terms of both runtime and efficiency. These results compellingly corroborate our model’s superiority in architectural image processing, enhancing measurement accuracy and drastically reducing execution time. [ABSTRACT FROM AUTHOR]

Algorithms for converting 2D to 3D are gaining importance following the hiatus brought about by the discontinuation of 3D TV production; this is due to the high availability and popularity of virtual reality systems that use stereo vision. In this paper, several depth image-based rendering (DIBR) approaches using state-of-the-art single-frame depth generation neural networks and inpaint algorithms are proposed and validated, including a novel very fast inpaint (FAST). FAST significantly exceeds the speed of currently used inpaint algorithms by reducing computational complexity, without degrading the quality of the resulting image. The role of the inpaint algorithm is to fill in missing pixels in the stereo pair estimated by DIBR. Missing estimated pixels appear at the boundaries of areas that differ significantly in their estimated distance from the observer. In addition, we propose parameterizing DIBR using a singular, easy-to-interpret adaptable parameter that can be adjusted online according to the preferences of the user who views the visualization. This single parameter governs both the camera parameters and the maximum binocular disparity. The proposed solutions are also compared with a fully automatic 2D to 3D mapping solution. The algorithm proposed in this work, which features intuitive disparity steering, the foundational deep neural network MiDaS, and the FAST inpaint algorithm, received considerable acclaim from evaluators. The mean absolute error of the proposed solution does not contain statistically significant differences from state-of-the-art approaches like Deep3D and other DIBR-based approaches using different inpaint functions. Since both the source codes and the generated videos are available for download, all experiments can be reproduced, and one can apply our algorithm to any selected video or single image to convert it. [ABSTRACT FROM AUTHOR]

The research introduces a unique deep-learning-based technique for remote rehabilitative analysis of image-captured human movements and postures. We present a ploninomial Pareto-optimized deep-learning architecture for processing inverse kinematics for sorting out and rearranging human skeleton joints generated by RGB-based two-dimensional (2D) skeleton recognition algorithms, with the goal of producing a full 3D model as a final result. The suggested method extracts the entire humanoid character motion curve, which is then connected to a three-dimensional (3D) mesh for real-time preview. Our method maintains high joint mapping accuracy with smooth motion frames while ensuring anthropometric regularity, producing a mean average precision (mAP) of 0.950 for the task of predicting the joint position of a single subject. Furthermore, the suggested system, trained on the MoVi dataset, enables a seamless evaluation of posture in a 3D environment, allowing participants to be examined from numerous perspectives using a single recorded camera feed. The results of evaluation on our own self-collected dataset of human posture videos and cross-validation on the benchmark MPII and KIMORE datasets are presented. [ABSTRACT FROM AUTHOR]

The research introduces a unique deep-learning-based technique for remote rehabilitative analysis of image-captured human movements and postures. We present a ploninomial Pareto-optimized deep-learning architecture for processing inverse kinematics for sorting out and rearranging human skeleton joints generated by RGB-based two-dimensional (2D) skeleton recognition algorithms, with the goal of producing a full 3D model as a final result. The suggested method extracts the entire humanoid character motion curve, which is then connected to a three-dimensional (3D) mesh for real-time preview. Our method maintains high joint mapping accuracy with smooth motion frames while ensuring anthropometric regularity, producing a mean average precision (mAP) of 0.950 for the task of predicting the joint position of a single subject. Furthermore, the suggested system, trained on the MoVi dataset, enables a seamless evaluation of posture in a 3D environment, allowing participants to be examined from numerous perspectives using a single recorded camera feed. The results of evaluation on our own self-collected dataset of human posture videos and cross-validation on the benchmark MPII and KIMORE datasets are presented.

Image interpolation is an essential process for image processing and computer graphics in wide applications to medical imaging. For image interpolation used in medical diagnosis, the two-dimensional (2D) to three-dimensional (3D) transformation can significantly reduce human error, leading to better decisions. This research proposes the type-2 fuzzy neural networks method which is a hybrid of the fuzzy logic and neural networks as well as recurrent type-2 fuzzy neural networks (RT2FNNs) for advancing a novel 2D to 3D strategy. The ability of the proposed methods in the approximation of the function for image interpolation is investigated. The results report that both proposed methods are reliable for medical diagnosis. However, the RT2FNN model outperforms the type-2 fuzzy neural networks model. The average squares error for the recurrent network and the typical network reported 0.016 and 0.025, respectively. On the other hand, the number of fuzzy rules for the recurrent network and the typical network reported 16 and 22, respectively.

Image interpolation is an essential process for image processing and computer graphics in wide applications to medical imaging. For image interpolation used in medical diagnosis, the two-dimensional (2D) to three-dimensional (3D) transformation can significantly reduce human error, leading to better decisions. This research proposes the type-2 fuzzy neural networks method which is a hybrid of the fuzzy logic and neural networks as well as recurrent type-2 fuzzy neural networks (RT2FNNs) for advancing a novel 2D to 3D strategy. The ability of the proposed methods in the approximation of the function for image interpolation is investigated. The results report that both proposed methods are reliable for medical diagnosis. However, the RT2FNN model outperforms the type-2 fuzzy neural networks model. The average squares error for the recurrent network and the typical network reported 0.016 and 0.025, respectively. On the other hand, the number of fuzzy rules for the recurrent network and the typical network reported 16 and 22, respectively. [ABSTRACT FROM AUTHOR]

• Statistical deformation and texture models (SDTMs) were created based on QCT scans. • 3D finite element models of human femora were reconstructed from 2D projections. • The reconstructed finite element models were validated using experimental data. Ultimate force of the proximal human femur can be predicted using Finite Element Analysis (FEA), but the models rely on 3D computed tomography images. Landmark-based statistical appearance models (SAM) and B-Spline transformation-based statistical deformation models (SDM) have been used to estimate 3D images from 2D projections, which facilitates model generation and reduces the radiation dose. However, there is no literature on the accuracy of SDM-based FEA models of bones with respect to experimental results. In this study, a methodology for an enhanced SDM with textural information is presented. The statistical deformation and texture models (SDTMs) are based on a set of 37 quantitative CT (QCT) images. They were used to estimate 3D images from two or one projections of the set in a leave-one-out setup. These estimations where then used to create FEA models. The ultimate force predicted by FEA models estimated from two or one projection using the SDTMs were compared to the experimental ultimate force from a previous study on the same femora and to the results of standard QCT-based FEA models. High correlations between predictions and experimental measurements were found for FEA models reconstructed from 2D projections with R 2 = 0.835 when based on two projections and R 2 = 0.724 when using one projection. The correlations were comparable to those reached with standard QCT-based FE-models with the experimental results (R 2 = 0.795). This study shows the high potential of SDTM-based 3D image reconstruction and FEA modelling from 2D projections to predict femoral ultimate force. [ABSTRACT FROM AUTHOR]

Histological brain slices are widely used in neuroscience to study the anatomical organization of neural circuits. Systematic and accurate comparisons of anatomical data from multiple brains, especially from different studies, can benefit tremendously from registering histological slices onto a common reference atlas. Most existing methods rely on an initial reconstruction of the volume before registering it to a reference atlas. Because these slices are prone to distortions during the sectioning process and often sectioned with non-standard angles, reconstruction is challenging and often inaccurate. Here we describe a framework that maps each slice to its corresponding plane in the Allen Mouse Brain Atlas (2015) to build a plane-wise mapping and then perform 2D nonrigid registration to build a pixel-wise mapping. We use the L2 norm of the histogram of oriented gradients difference of two patches as the similarity metric for both steps and a Markov random field formulation that incorporates tissue coherency to compute the nonrigid registration. To fix significantly distorted regions that are misshaped or much smaller than the control grids, we train a context-aggregation network to segment and warp them to their corresponding regions with thin plate spline. We have shown that our method generates results comparable to an expert neuroscientist and is significantly better than reconstruction-first approaches. Code and sample dataset are available at sites.google.com/view/brain-mapping.

Accurate 3D measuring systems thrive in the past few years. Most of them are based on laser scanners because these laser scanners are able to acquire 3D information directly and precisely in real time. However, comparing to the conventional cameras, these kinds of equipment are usually expensive and they are not commonly available to customers. Moreover, laser scanners interfere easily with each other sensors of the same type. On the other hand, computer vision-based 3D measuring techniques use stereo matching to acquire the cameras' relative position and then estimate the 3D location of points on the image. Because this kind of systems needs additional estimation of the 3D information, systems with real time capability often relies on heavy parallelism that prevents implementation on mobile devices. Inspired by the structure from motion systems, we propose a system that reconstructs sparse feature points to a 3D point cloud using a mono video sequence so as to achieve higher computation efficiency. The system keeps tracking all detected feature points and calculates both the amount of these feature points and their moving distances. We only use the key frames to estimate the current position of the camera in order to reduce the computation load and the noise interference on the system. Furthermore, for the sake of avoiding duplicate 3D points, the system reconstructs the 2D point only when the point shifts out of the boundary of a camera. In our experiments, we show that our system is able to be implemented on tablets and can achieve state-of-the-art accuracy with a denser point cloud with high speed. [ABSTRACT FROM AUTHOR]

In molecular sciences, articles tend to revolve around 2D representations of 3D molecules, and sighted scientists often resort to 3D virtual reality software to study these molecules in detail. Blind and visually impaired (BVI) molecular scientists have access to a series of audio devices that can help them read the text in articles and work with computers. Reading articles published in this journal, though, is nearly impossible for them because they need to generate mental 3D images of molecules, but the article-reading software cannot do that for them. We have previously designed AsteriX, a web server that fully automatically decomposes articles, detects 2D plots of low molecular weight molecules, removes meta data and annotations from these plots, and converts them into 3D atomic coordinates. AsteriX-BVI goes one step further and converts the 3D representation into a 3D printable, haptic-enhanced format that includes Braille annotations. These Braille-annotated physical 3D models allow BVI scientists to generate a complete mental model of the molecule. AsteriX-BVI uses Molden to convert the meta data of quantum chemistry experiments into BVI friendly formats so that the entire line of scientific information that sighted people take for granted-from published articles, via printed results of computational chemistry experiments, to 3D models-is now available to BVI scientists too. The possibilities offered by AsteriX-BVI are illustrated by a project on the isomerization of a sterol, executed by the blind co-author of this article (HBW). [ABSTRACT FROM AUTHOR]

The characterization of 3D structures in porous media is crucial for predicting physical properties in many industries, such as CO2 capture and storage, hydrology, oil & gas. In contrast to the expensive and time-consuming acquisition of 3D images, 2D imaging can provide cheap and fast data. However, the reconstruction of a 3D image from a single 2D image is a complex non-deterministic inverse problem. Several statistical and deep learning-based algorithms have been introduced in the past, however, most of them fail to generalize structures and textures for different types of rocks, in addition to being time-consuming and only able to generate relatively small images (30 0 3 voxels cube). In this work, we propose a size-invariant multi-step 3D generation workflow from a single 2D image using a combination of Vector-Quantized Variational AutoEncoder(VQ-VAE), size-invariant Generative Adversarial Networks(GAN), and Image Transformer. The proposed workflow tackles several major challenges in the generation of 3D images since it is designed to not only satisfy the large size constraint (> 100 0 3 voxels cube) but also to generate statistically representative pore structures. The combination of these different generative techniques allows us to overcome the scalability, stability, and complexity associated with GAN approaches. We trained the proposed workflow using several types of rocks with different physical properties, sizes, and resolutions. To validate our methodology, we have generated several large-size 3D rock images and compare them to real 3D images in terms of physical properties (porosity, permeability, and Euler characteristic). • Single model for multiple types of rock using Transformer, VQ-VAE, and GAN. • Easy and stable to train, deploy, retrain and maintain. • Generate statistically representative 3D digital rock images from a single 2D image. • Generate images > 100 0 3 voxels of different rock types including carbonates. [ABSTRACT FROM AUTHOR]

Shape transformation of flat two-dimensional (2D) materials into three-dimensional (3D) complex shapes is a novel technology which can be explored to customize materials with attractive properties and reduce package space. This study aims at designing cereal-based xerogel discs, explore its application as a flat 2D product that can transform from 2D to 3D shape and characterize its hydromorphic behavior. The flat 2D xerogel disc was developed by preparing wheat hydrogel (10%) and drying it in sessile drop form at ambient conditions to get flat wheat xerogel discs. For programming shape transformation, ethyl cellulose strips were applied over the top surface of xerogel discs in different patterns. Shape programmed xerogel disc was immersed in hot water and shape transformation was recorded. Shape transformation was characterized by evaluating the parameters such as bending curvature, height, angle, and swelling power. As the shape transformation proceeds, an increase in bending curvature, angle, height, and end-to-end curve distance was observed with time. The overall folding time was found to be 50–60 seconds depending upon the constraint material pattern. This work provides a simple approach to design a flat 2D food product capable of changing its shape into 3D. [ABSTRACT FROM AUTHOR]